Dokploy

n8n Queue Mode

n8n running in queue mode for high-throughput production workloads: the main instance handles the editor and webhooks while a scalable pool of workers executes workflows from a Redis (Bull) queue, backed by PostgreSQL.

n8n Queue Mode logo

Configuration

services:
  postgres:
    image: postgres:17-alpine
    restart: unless-stopped
    environment:
      - POSTGRES_USER=${POSTGRES_USER}
      - POSTGRES_PASSWORD=${POSTGRES_PASSWORD}
      - POSTGRES_DB=${POSTGRES_DB}
    volumes:
      - postgres_data:/var/lib/postgresql/data
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER} -d ${POSTGRES_DB}"]
      start_period: 30s
      interval: 10s
      timeout: 5s
      retries: 5

  redis:
    image: redis:7-alpine
    restart: unless-stopped
    command: redis-server --requirepass ${REDIS_PASSWORD}
    environment:
      - REDIS_PASSWORD=${REDIS_PASSWORD}
    volumes:
      - redis_data:/data
    healthcheck:
      test: ["CMD-SHELL", "redis-cli -a \"$$REDIS_PASSWORD\" ping | grep PONG"]
      start_period: 20s
      interval: 10s
      timeout: 5s
      retries: 5

  # Main instance: serves the editor UI and webhooks, publishes executions
  # to the Redis (Bull) queue instead of running them itself.
  n8n:
    image: n8nio/n8n:2.30.4
    restart: unless-stopped
    environment:
      # PostgreSQL
      - DB_TYPE=postgresdb
      - DB_POSTGRESDB_HOST=postgres
      - DB_POSTGRESDB_PORT=5432
      - DB_POSTGRESDB_DATABASE=${POSTGRES_DB}
      - DB_POSTGRESDB_USER=${POSTGRES_USER}
      - DB_POSTGRESDB_PASSWORD=${POSTGRES_PASSWORD}

      # Queue mode (Redis / Bull)
      - EXECUTIONS_MODE=queue
      - QUEUE_BULL_REDIS_HOST=redis
      - QUEUE_BULL_REDIS_PORT=6379
      - QUEUE_BULL_REDIS_PASSWORD=${REDIS_PASSWORD}
      - QUEUE_HEALTH_CHECK_ACTIVE=true
      - OFFLOAD_MANUAL_EXECUTIONS_TO_WORKERS=true

      # Encryption key shared between the main instance and the workers
      - N8N_ENCRYPTION_KEY=${N8N_ENCRYPTION_KEY}

      # Networking
      - N8N_HOST=${N8N_HOST}
      - N8N_PORT=5678
      - N8N_PROTOCOL=http
      - N8N_PROXY_HOPS=1
      - NODE_ENV=production
      - N8N_WEBHOOK_URL=https://${N8N_HOST}/
      - N8N_EDITOR_BASE_URL=https://${N8N_HOST}/
      - GENERIC_TIMEZONE=${GENERIC_TIMEZONE}
      - N8N_SECURE_COOKIE=false
    volumes:
      - n8n_data:/home/node/.n8n
    depends_on:
      postgres:
        condition: service_healthy
      redis:
        condition: service_healthy
    healthcheck:
      test: ["CMD-SHELL", "wget --spider -q http://127.0.0.1:5678/healthz"]
      start_period: 60s
      interval: 10s
      timeout: 5s
      retries: 10

  # Queue workers: pull executions from Redis and run them. Scale them by
  # changing the N8N_WORKER_REPLICAS environment variable and redeploying.
  n8n-worker:
    image: n8nio/n8n:2.30.4
    restart: unless-stopped
    command: worker
    deploy:
      replicas: ${N8N_WORKER_REPLICAS:-2}
    environment:
      # PostgreSQL
      - DB_TYPE=postgresdb
      - DB_POSTGRESDB_HOST=postgres
      - DB_POSTGRESDB_PORT=5432
      - DB_POSTGRESDB_DATABASE=${POSTGRES_DB}
      - DB_POSTGRESDB_USER=${POSTGRES_USER}
      - DB_POSTGRESDB_PASSWORD=${POSTGRES_PASSWORD}

      # Queue mode (Redis / Bull)
      - EXECUTIONS_MODE=queue
      - QUEUE_BULL_REDIS_HOST=redis
      - QUEUE_BULL_REDIS_PORT=6379
      - QUEUE_BULL_REDIS_PASSWORD=${REDIS_PASSWORD}
      - QUEUE_HEALTH_CHECK_ACTIVE=true

      # Must be the same key as the main instance
      - N8N_ENCRYPTION_KEY=${N8N_ENCRYPTION_KEY}

      - NODE_ENV=production
      - GENERIC_TIMEZONE=${GENERIC_TIMEZONE}
    depends_on:
      n8n:
        condition: service_healthy
    healthcheck:
      test: ["CMD-SHELL", "wget --spider -q http://127.0.0.1:5678/healthz"]
      start_period: 60s
      interval: 10s
      timeout: 5s
      retries: 10

volumes:
  postgres_data:
  redis_data:
  n8n_data:
[variables]
main_domain = "${domain}"
postgres_user = "${username}"
postgres_password = "${password:24}"
postgres_db = "n8n"
redis_password = "${password:24}"
n8n_encryption_key = "${base64:32}"

[config]
mounts = []

[[config.domains]]
serviceName = "n8n"
port = 5_678
host = "${main_domain}"

[config.env]
N8N_HOST = "${main_domain}"
GENERIC_TIMEZONE = "Europe/Berlin"

# Number of queue workers. Change it and redeploy to scale.
N8N_WORKER_REPLICAS = "2"

# PostgreSQL
POSTGRES_USER = "${postgres_user}"
POSTGRES_PASSWORD = "${postgres_password}"
POSTGRES_DB = "${postgres_db}"

# Redis (Bull queue)
REDIS_PASSWORD = "${redis_password}"

# Encryption key shared by the main instance and the workers (IMPORTANT)
N8N_ENCRYPTION_KEY = "${n8n_encryption_key}"

Base64

To import this template in Dokploy: create a Compose service → Advanced → Base64 import and paste the content below:

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

Instructions

n8n Queue Mode

This template runs n8n in queue mode:

  • n8n (main): serves the editor UI and receives webhooks/triggers, then publishes executions to the queue.
  • n8n-worker (x2 by default): pulls executions from the Redis (Bull) queue and runs them.
  • redis: the message broker for the queue (password protected).
  • postgres: stores workflows, credentials and execution data.

All n8n containers share the same N8N_ENCRYPTION_KEY (generated automatically), which is required for workers to decrypt credentials.

Scaling workers

Worker replicas are controlled by the N8N_WORKER_REPLICAS environment variable (default 2):

  1. Open the compose service in Dokploy and go to the Environment tab.
  2. Change N8N_WORKER_REPLICAS to the number of workers you want.
  3. Redeploy. Docker Compose will scale the n8n-worker service to the requested number of replicas.

Each worker runs up to 10 concurrent executions by default. You can tune this by adding N8N_CONCURRENCY_PRODUCTION_LIMIT to the worker environment.

Optional: dedicated webhook processors

For very high webhook volumes, n8n supports dedicated webhook processor instances (command: webhook). To use them, add another service to the compose file with the same environment as n8n-worker but with command: webhook, and route the /webhook/ path of your domain to it. See the n8n queue mode docs for details.

Tags

automation, workflow, low-code, queue

Version: 2.30.4

n8n

n8n is an open source low-code platform for automating workflows and integrations.

n8n + Worker + Runner with Redis/Postgres and Ollama

n8n is an open source low-code platform for automating workflows and integrations with PostgreSQL database and Ollama AI model.

On this page

ConfigurationBase64Instructionsn8n Queue ModeScaling workersOptional: dedicated webhook processorsLinksTags